Japanese-language iPaq

A user in Japan reported that random Japanese characters were appearing in the Gospel display when she was using the stand-alone Universalis on her Japanese-language iPaq.

This was because of a difference in the way that Microsoft’s operating system (Windows Mobile 5.0) works on international and Japanese-language palmtops. The stand-alone (downloadable) version of Universalis has been altered to work round this problem, and the version now available for download from the Universalis site will work identically on Windows Mobile computers anywhere in the world.

For technical readers, and for Google searchers who are looking for the reason for this strange behaviour: in the rest of the world, the CP_ACP option of MultiByteToWideChar uses a single-byte character set (usually Windows 1252). In Japan (and probably in other East Asian countries) it uses a double-byte character set. This behaviour is poorly documented. It needs to be worked round if you have characters such as quotation marks in the text you are trying to display, because the quotation mark will be taken as the first byte of a DBCS character.